| dc.description |
This study focuses on the nasalization of the postposed determiner /la/ (LÃ) after an oral segment (e.g. chat la/lã [ʃatla/lã] ‘the cat’, and peyi a/an [pejija/ã), a linguistic environment where the nasal variants generally do not occur. In his 1991 pilot study, Valdman demonstrated that there was a correlation between younger middle-class Port-au-Prince speakers and the nasalization of the determiner when following an oral segment. I used a variationist sociolinguistic approach to investigate the issue more extensively and to provide substantive answers to three research questions: (1) Has this linguistic change extended to other social groups, for example, to monolingual speakers of Haitian Creole? (2) Are there linguistic factors conditioning the change, for example, the phonological features of vowels in word-final syllables? (3) Is there a correlation between Frenchified features (e.g. front rounded vowels, postvocalic [r]) and the nasalization of the determiner in non-nasal environments? The corpus includes three sets of data gathered from pair interviews (P), individual interviews, (I) and data elicitation (E) conducted with 32 natives of Haitian Creole. The speakers’ social profiles were coded for age, sex, geographical location, occupation, education and level of bilingualism. The results show that the nasalization of the determiner LA in non-nasal contexts has been extended to speakers of different social status, particularly to monolingual speakers as well as those living in different geographical areas of the country (i.e. urban and rural). Regarding the vi effect of linguistic environments, the results reveal that high vowels favor LÃ across the board. However, LÃ does not occur with low vowels in open syllables as a result of vowel lengthening, which then blocks vowel nasalization (e.g. papa a [papa:]/*papa an [papaã] ‘the father’). Finally, the study indicates a link between Frenchified features and nasalization of the determiner for some speakers and not for others. Even though Frenchified features occurred less frequently among the monolingual speakers, those with average or higher level of education nasalized the determiner more than their peers when these features were present. |
|